cz sk en

Regular KPI Performance Monitoring


To achieve a return on investment in an e-commerce project, it is necessary to monitor the Key Performance Indicators (KPI). These will tell us if the investment is paying off and what potentially needs to be done to improve the situation. We looked at how to become familiar with a number of indicators and how to get a picture of the real state of the company.

In this article we will frequently use the terms KPI, goal and metrics. Let’s define what we mean by each of them.

  • Goal: the state that we want to achieve
  • Metrics: any indicator that we can express numerically
  • KPI: Key Performance Indicator, i.e. the most important metrics being monitored

5 reasons to regularly monitor KPI

There is so much data in marketing that it is often difficult to separate important trends from ordinary noise. It’s easy to get lost among the number of impressions, clicks, page visits and conversions. Each metric has its own justification, but one has to distinguish between the knowledge that it brings us and the context. By themselves these numbers don’t really tell us a lot. Partial KPIs are, however, useful aids for the attentive follower, and monitoring them regularly can tell us a lot. 

  1. They inform us about the current status
  2. They enable us to set goals and monitor their fulfilment
  3. Thanks to them we can anticipate trends over time
  4. They help us learn and move forward
  5. They bring tips for making changes and improvements

With correct KPI monitoring, we should be able to assess the situation minimally in the form we are doing well/we are doing badly. If we do not monitor KPIs regularly or at all, we may miss various types of essential information that can help our e-shop go forward. 

Classification of KPIs according to their impact

It is first necessary to determine the impact of the metric on the overall business. The classification of metrics from the well-known expert in online marketing, Avinash Kaushik, is very helpful:

Classification of metrics from the well-known online marketing expert Avinash Kaushik

This graph describes on the x-axis what time frame of the metric is important to us. The y-axis shows the level of the decision that the given metric influences. While among the data that needs to be monitored every week we find, for example, conversions or Bounce rates, in the long-term view we should not forget to monitor the probability of a recommendation or Customer Lifetime Value. Not all metrics are listed in the graph, of course; they will probably be different for each type of business. But this does give us an idea of how to classify and use them wisely.

In order to translate this into the common life of a company: the number of clicks or likes is not important for a financially successful month. These describe only the immediate reaction of people to the advertising message. In contrast, in the given context, we will be interested in the volume of turnover or margin, or the purchase rate and turnover per visit.

Don’t forget the right KPI for the right role

Just as we differentiate the level of metrics for further decision-making (operational up to strategic), we must likewise perceive the context of the metrics for each position of people in the company. From the point of view of the project investor, it’s not important how many clicks the ads got or the number of visits made to the website. At this level, KPIs are more important than turnover, profitability, long-term customer value and, obviously, their development over time.

Understandable for the investor

For useful communication between an investor and a company’s CEO or marketing manager (CMO), it is specifically important to understand how important the given metric is for the other and what we want to communicate with it. This follows from what each person simply experiences in their role. 

Let’s see how it looks in practice. If a CMO daily addresses some topic with his team, it is not easy for him to suddenly change his thinking and put himself in the role of CEO, for example, and present the status and marketing plan so that it is comprehensible and useful (i.e. actionable) for the CEO. I have personally experienced several situations in which, for example, an investor did not fully understand what the marketing manager was trying to present to him. In this case, it is more the head of marketing who must present himself such that he clearly connects his activities with the final results of the company. On the other hand, for an investor who may be an expert in one or several areas outside of marketing, it is difficult to suddenly know the complexity of the marketing strategy of a company he is investing in. It can then be difficult for him to assess whether the company’s marketing strategy is set up correctly, or what steps are necessary to take to improve the situation.

It is this bridge between the investor and the CMO (or the CEO presenting the marketing results) specifically that an independent consultant can help with. Firstly, as a kind of translator between different KPIs, as someone who understands the investment and marketing perspective, and also as a watchdog, who regularly monitors marketing performance and can quickly and specifically point to what needs to be changed.

Sources of data and their connections

Since we have looked at the concept of how individual KPIs fit into the context of decision-making levels and individual roles of people in the company, let’s take a practical look at what data sources an e-shop typically has and how we can connect and use them. The most common source of data for the e-shop are: 

  • Its own system (an ERP or CMS system with database of products and information about sales, finances and warehouse) – this is the most accurate source of data, but it often lacks “soft” metrics on web behaviour or performance from marketing systems.
  • Web Analytics – Google Analytics, Exponea, or another web analytics solution – this source of data well captures user behaviour on the web, but it may not be accurate in terms of financial results and sales. It does not record 100% of orders and with standard settings does not contain return and margin information.
  • Marketing platform tools such as Google Ads or Facebook Business Manager. These contain in particular data on the impact and performance of ads and the minutes of credit before entering a site.
  • Other sources of data, such as accounting, warehousing or logistics systems, can provide, for example, data on product purchase prices, price and length of transport, the value currently tied up in the warehouse, or fixed costs.

With the gradual growth of the e-shop, it is also necessary to monitor the amounts of other data. In our article Technological background of e-shops during growth, you can also learn more about various auxiliary tools for the transition to automation.

We can here look at data sources from different perspectives; in the range of products, channels, and users (indicated by our customer’s long-term value metrics). Let’s now look at these individual optics in more detail. 

Product profitability

Let’s try to imagine that the only source of income for an e-shop is the sale of products (let’s ignore various other sources of income). In such a case, all revenues, costs and profits of the company can be calculated for one product.

The first level that every e-shop needs to handle is to have an overview of the purchase and sale price of its products. In this way we will find out the gross margin for each product. This is important information for prioritising which products are more worth selling and for calculating net profit per product. After calculating the gross margin per product, we can then calculate additional variable and fixed costs that we need to calculate for a single product.

Do you know the difference between a margin and a markup and how to calculate the two of them? For every e-shop, the correct analysis of operations costs, calculation of the cost level and the marketing percentage is also very important.

Channel profitability

Aside from products, marketing channels need to be managed effectively. Each channel requires some investment, whether directly (such as click credits) or minimal internal administration costs. The source of information on investment and performance is most often at the interface of the given channel, i.e. at the interface of Facebook Business Manager or Google Ads, Heureka, or an order for a television spot. Another often used source of information is Google Analytics, or some other analytics tool applied on the web. This can provide data on the behaviour of users after their arrival from a particular channel.

A limitation of analytical channels, such as Google Analytics, is their dependence on cookies, which are not completely reliable in identifying users across devices or over time. At the same time, with the use of several marketing channels, the question of attribution arises – that is, which of the channels was the most important, which decided on the purchase or other important act. This is particularly difficult with purely visual formats, such as television advertising, where we are virtually blind to the period between seeing an ad and making a purchase. Various tools, such as Causal Impact or other applications for advanced attribution models, can help with these limitations. The Causal Impact tool is available as an R package from Google.

Long-term customer value

Measuring the profitability of products and channels often takes place at one point in time. But this, too, is limiting. In reality, users are gradually divided into repeat users and new users. What’s more, each industry has a different purchase cycle length. I’ll mention the furniture segment versus nutritional supplements as an example. Each of these has a different length of decision-making and frequency of repeat purchases. Therefore, Customer Lifetime Value is an extremely important metric and is key for determining marketing investment.

For a simple calculation of Customer Lifetime Value (CLV), Google Analytics (based on a cookie) is sufficient for us. The most reliable source of data for the CLV calculation comes directly from the internal system (e.g. ERP), where the customer’s personal data is stored, as is data on when and how often he made a purchase. Furthermore, we are not bothered by cookie restrictions here because we have complete first-party data.

Using CLV is key when determining your marketing investment, but it also allows us to categorise products well. Some products may be loss-making, but if they are introductory products that attract new customers and turn them into regular customers, the value of such a product for the e-shop is suddenly completely different.

How many KPIs to monitor?

At the beginning, select just a few KPIs. Many people think that the more indicators they monitor, the more results they will get. But paradoxically, a large number of monitored KPIs can lead to confusion and chaos in achieving goals. Therefore, we recommend at the beginning that you choose only 2–3 indicators for a specific role and pay more attention to them. 

Data visualisation

The collecting and connecting of data is not the main goal in itself. It is much more important to gain knowledge from the data and use it to make the right decisions in the future. Especially for management positions in the company and for investors, it is important to have up-to-date and exact information describing the status and development of the e-shop. For the best possible presentation and interpretation of data, we must correctly visualise them.

In the second section, we discussed the importance of the right KPIs for different roles in the company. The same applies to the correct visualisation of data and their presentation. For more strategic roles and investors in particularly, it is important not to overwhelm them with irrelevant metrics but to clearly communicate only the most important things, because there is often little time for presentation. On the other hand, more operative roles in the e-shop can make good use of live dashboards on the current performance of trade or marketing, or on selected channels. Opinest can help you choose the range and suitability and the kind data to present to whom and in what way.

Visualisation tools

These are among the most commonly used visualisation tools that we recommend. 

Google Data Studio – easy to use, but it does not have much variability, good connection with other Google products

Microsoft Power BI – a more sophisticated tool with good links to other Microsoft tools (such as Excel)

Tableau – this is probably the market leader in visualisation tools, but using it can be expensive

Other widely used tools include, for example, Infogram, Zoho Analytics or Datawrapper

For a more detailed understanding of specific KPI indicators or the creation of the already mentioned bridge between an investor and an CEO or CMO, an independent consultant is a brilliant choice. If you are also considering this or have more questions about KPIs, you are in the right place. Feel free to write to us


Are you ready for rocket growth?

Contact us

Book a call

Choose person:

Matej Karaba

Long-term Impact & Business Consultant

Michal Lubelec

Consultant for E-commerce Analytics and Measurement

Marek Ďuračka

Social Media Consultant

  • Matej Karaba

    Long-term Impact & Business Consultant

  • Michal Lubelec

    Consultant for E-commerce Analytics and Measurement

  • Marek Ďuračka

    Social Media Consultant

Matej is Long-term Impact & Business Consultant and will help you with:

  • Coverage of the marketing mix potential
  • Long-term sustainability
  • Development of a business strategy
  • Creativity in technology
  • Managing IT projects
  • UX/UI and SEO

Michal is Consultant for E-commerce Analytics and Measurement and will help you with:

  • data and analytics settings
  • bidding and budget planning
  • campaign automation

Marek is Social Media Consultant will help you with:

  • Facebook
  • Instagram
  • ROI Hunter